Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!uwm.edu!dogie.macc.wisc.edu!uakari.primate.wisc.edu!brutus.cs.uiuc.edu!apple!jrg
From: jrg@Apple.COM (John R. Galloway Jr.)
Newsgroups: comp.sys.mac.hardware
Subject: Re: Mac IIfx or IIci (parity)
Message-ID: <39856@apple.Apple.COM>
Date: 27 Mar 90 08:19:43 GMT
References: <18876@boulder.Colorado.EDU> <22291@netnews.upenn.edu>
Distribution: usa
Organization: Galloway Research
Lines: 36

In article <22291@netnews.upenn.edu> meuchen@grad1.cis.upenn.edu (Paul Eric Menchen) writes:
>In article <18876@boulder.Colorado.EDU> watermaa@tramp.Colorado.EDU (WATERMAN ALEXANDER S) writes:
>>
>>Is it [parity] worth the extra $450 for this feature?
>>
>>Kent Leung
>Do you use parity checking for your modem transmissions or 8-N?
>Given the choice you would probably want no parity. Most poeple have
>now agreed that parity is a bad way to check for errors. Have you ever
>had a memory error, besides the power being turned off? I doubt it.

   I suspect your are right (as you state later) that it was for gov bids.
But your analogy seems syspect.  If you are running any sort of file xfer
on your modem, then there IS error checking on a block basis at the higher
level. If runing interactively you will (or at least can) see the error when
a character is echoed incorrectly.  But there is no higher level when you
load and execute a program or load a database.  Would you turn ERCC checking
off on your disk? Likey not (even if you could).  I have seen parity errors,
very rarely but it seems like cheap protection especially if you are using
your machine for vital informaiton (like your accounts or checkbook etc.).
I would rather have a longer ECC code on longer data (i.e. on a 32/64/128 bit
item) but if parity is all I can get, I think I would tend to take it.
ALso parity errors become more likely as DRAMs get denser and hence the bit
cell holds fewer electrons and is more easily flipped.  Perhaps the best
solution are DRAMs (available I think) that have internal error detection and
correction.  That way the only part of the system that has to worry about
the issue, is that which is most vulnerable, and the rest need not be concerned.
  I am POed that Apple bundled the parity option with the 80SC, so you can't
get a diskless IIfx with parity.
	-jrg
-- 
internet   jrg@apple.com      John R. Galloway, Jr.
applelink  d3413              CEO..receptionist         795 Beaver Creek Way
human     (408) 259-2490      Galloway Research         San Jose, CA  95133

These are my views, NOT Apple's, I am a GUEST here, not an employee!!